Some Early Thoughts on Researching Democratic Grantmaking

In autumn, I will begin my doctoral studies at the University of Michigan School of Information as a STIET Fellow. This is a great opportunity for me to research how democratic media – media that structurally and procedurally incorporates democratic principles such as inclusiveness, transparency, and equal power sharing – influences microeconomic behavior. It gives me an opportunity to research some basic questions about the economic efficiency of democratic culture, and how to improve it.

How would we actually allocate limited shared resources, like our pooled tax money or pooled charitable donations, if we actually were permitted to directly discuss, evaluate, and vote on those allocations? We don’t really know because instead of directly participating in collective-decision making processes to allocate our shared resources, most of us mainly rely on a relatively small group of corporate leaders from government, industry, and the nonprofit sector to make those decisions for us.

Part of the reason why we do not directly participate in collective-decision making processes to allocate our shared resources is because we lack the information infrastructure to do so. For example, our banking websites do not allow us to set up group accounts and collectively vote on how our money is spent. Our government websites, even at the local level, do not allow us to collectively propose, rate or review public budget items. There are many websites that allow us to directly donate funds to nonprofit organizations, but almost none allow us to vote on how those charitable funds, once collected, are spent.

In each case, we the stakeholders are mainly viewed as funders (investors, taxpayers, donors), but not as deciders (executives, legislators, philanthropists). We are largely viewed as a raw source of money and labor for the organizations that purport to represent us, not as the democratic leadership that designs the policies and signs the checks for those organizations. So we need to research whether there are any economically viable democratic alternatives, whether there are any democratic information models and interfaces that can incentivize more fair and efficient allocations of shared resources. Perhaps the status quo, with its veneer of representative democracy, is the most fair and efficient socio-technical infrastructure for allocating our shared resources, but we do not know; we have not experimentally explored many alternatives.

So in general, I’m interested in researching whether democratic media can be used to incentivize fair and efficient allocations of organizational monies. I’m interested in researching the fairness and efficiency of democratic resource allocation mechanisms, especially those that allocate organizational monies in the form of minigrants, and the informational factors that contribute to their fairness and efficiency.

Another major reason why we have not directly participated in collective decision-making processes to allocate our shared resources is because of general arguments against direct democracy, which we occasionally hear and incuriously accept, especially the argument that we cannot use direct democracy to solve our resource allocation problems because solutions require expertise which we lack, which we cannot or will not learn, or which we cannot or will not share. This argument assumes that the status quo has and makes better use of expertise than every potential model of direct democracy, and that the current corporate leadership have and use more expertise than we have or could use.

However, I am not convinced that our politicians and CEOs either 1) have or use expertise more efficiently than every other potential group of voters, or 2) that every model of directly democratic decision-making is bound to lack sufficient expertise and participation. For example, most of our corporate deciders are lawyers, businessmen, or community organizers – not universal policy experts. Like every human being, they have major gaps in knowledge and rely on others to advise them during decision making processes. So why shouldn’t the experts that advised our politicians and CEOs on particular issues also have direct influence on the decision-making process of those issues? Why must their expertise be filtered by some minority of lawyers, businessmen, and community organizers?

In fact, it seems quite feasible, and potentially more fair and efficient, to create an online space where the public, including experts from all walks of life, can directly propose, review, and vote on how to allocate organizational monies, especially in the context of community minigrants.

If a person does not feel like they have the time or expertise, or if they are unwilling or unable, to directly vote on every minigrant, they can temporarily delegate or “loan” their vote for specific types of grants to other users that they trust and perceive to be more expert on those types of grants than themselves. Moreover, by default, people could delegate their votes to the established corporate leadership (this would be a vote for the status quo and should be available).

There are several other design features which would make this model of direct democractic decision-making more robust, such as the ability to reclaim and recast your vote prior to the vote’s deadline, and the ability for those who are currently stewards of loaned votes, to be able to reloan those votes to even more trusted experts in their social networks, thereby creating chains of loaned votes. The stewards, or representatives, higher up in the chain should have the ability to retrieve and recast the votes they loaned to any steward lower down on the chain. Of course, the owner of each vote would be the highest point in the chain, and she would have the ability to reclaim and recast her vote at anytime prior to the vote’s deadline from any steward of her vote. With this model, I predict that votes would tend to be dynamically routed and loaned along a chain of users from the owners of the votes to the locally perceived experts to the globally perceived experts, thereby increasing the overall satisfaction with the allocation of organizational monies.

This system of chained vote loaning would preserve the possibility of direct democracy, while encouraging the information efficiencies of representative democracy. It would significantly mitigate the threat of inherent information asymmetries from interfering with informed decision making. It would encourage greater political participation and knowledge sharing. It would advance personal responsibility for political and economic outcomes. Perhaps, most importantly, if adopted more broadly by government and industry, it would radically undermine the economic and mental waste poured into political ad campaigns. Since everyone is an alternative candidate, it would be too expensive to demonize every alternative representative. Without as many central characters or simple teams to glorify and demonize, news companies would find it more difficult to profit from horse-race non-sense; they would instead have to focus more on the cost-benefit analysis of policy proposals; they would have to get back to the job of informing our democracy.

Unlike the accountability and transparency systems currently found in government legislatures, if the people decided that their putative global experts were in fact ill-informed pundits, our model of chained vote loaning would allow them to immediately redistribute their votes to other trusted global experts, thereby dynamically re-percolating power to more trustworthy representatives. With the proper information infrastructure, this could reasonably occur because the owner of each vote would be able to see who is currently representing her on each issue and how her vote is currently cast. If dissatisfied with her current representative, she would be able to reclaim her vote and recast it, or reloan it to another user that better represents her individual interests.

I have only offered an adumbrated account of my model for chained vote loaning, but one worth researching and experimentally testing, especially in the context of community minigrants. I am very interested in determining how chained vote loaning compares to traditional models of representation, such as collectively elected representatives or appointed representatives, with respect to stakeholder participation and satisfaction.

Finally, I am generally interested in how recommender systems and other forms of social computing can facilitate democratic decision making. To make informed decisions over a large set of proposals, we need a way to democratically discover proposals and potential representatives. If our collaborative filtering mechanisms only attempt to account for the preferences of uninformed users, they could potentially only suggest the adoption of proposals supported by similarly uninformed users, which could snowball into the general adoption of uninformed proposals. What is needed is a way to incorporate the preferences of representative experts, while ultimately honoring the preferences of uninformed users. With chained vote loaning, one option would be to incorporate the preferences of a vote owner with her chained representatives’ preferences.

If pressed for a somewhat short list of research interests, I’d provide the following:

• democratic media
• incentive-centered design
• social informatics
• information economics
• experimental economics
• social choice theory
• social finance
• economic sociology
• market design
• social computing
• recommender systems

---